TSMC is laying out a clear path for its next wave of chip technology, outlining what’s coming over the next few years at its Open Innovation Platform Ecosystem Forum.
The company has already moved into full production of its N2 (2nm) process this year, marking its shift from FinFET to Nanosheet transistors. AMD is among the earliest adopters, using the new node for its Zen 6 CPU chiplets, which are expected to reach the market in 2026.
TSMC isn’t stopping at standard N2. Several follow-up versions are planned, including N2P and N2X, both designed to enhance efficiency and performance.
Another branch, A16 SPR, uses a “Super Power Rail” design that mirrors Intel’s backside power delivery approach. Early figures reported by HardwareLuxx suggest that A16 can run 8–10% faster than N2P at the same voltage or cut power consumption by 15–20% at equal performance.
After the N2 family, TSMC’s next significant leap will be the A14 node, which promises even tighter transistor density, along with improved speed and energy efficiency.
Steady gains in power and efficiency
TSMC expects both N2 and A14 to continue pushing efficiency forward. The company forecasts that A14 could offer up to 1.8× higher performance at the same power level and over four times better power efficiency. Historically, each new node has reduced power needs by around 30%, although recent charts suggest that these efficiency gains may be slowing somewhat.
Performance scaling, however, appears to be holding steady. N2 is projected to outperform N3E by approximately 15%, while A14 is expected to improve on N2 by roughly 16%. In short, TSMC expects to keep delivering faster transistor designs as it moves through its upcoming nodes.
For the wider tech industry, the takeaway is encouraging: TSMC’s roadmap remains on schedule, and upcoming nodes continue to promise gains in density, speed, and energy efficiency. Faster and more efficient processors are still very much on the horizon.
